[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#921226: RFS: austin/0.6.1~beta-1 [ITP]



Hi

Sorry for not including the bug address in the CC. I have simplified
the debian/rules file
(https://github.com/P403n1x87/austin/commit/655c3beb09199bb9dbe6c9dbdb8395c25e9213c9)
and verified with gbp buildpackage. I have excluded the custom
Makefile from the sources and seems fine to me now.

Please let me know if you want me to re-upload the package with the
latest changes.

Regarding python2, the package doesn't include any python2 code. Even
the python project inside austin/ is compatible with python3. However,
the main C application shipped with the package (which is just the
single binary src/austin) supports both python2 (2.3 - 2.7) and
python3 (3.3 - 3.7).

Thanks,
Gabriele

On Sun, 3 Feb 2019 at 23:31, Adam Borowski <kilobyte@angband.pl> wrote:
>
> On Sun, Feb 03, 2019 at 10:39:09PM +0000, Gabriele wrote:
> > Hey
> >
> > Many thanks for getting back to me!
>
> Could you please re-add 921226@bugs.debian.org to CC?  (I'm 99.9% sure
> you'll be ok with that but the rule is to never make anything public without
> an explicit permission).  Any such package reviews are supposed to be done
> in public, for a number of reasons:
>  * there's a record of the review
>  * anyone else can chime in
>  * anyone else can continue the review -- I might need to pass if there'll
>    be any complex python-specific questions
>  * others can learn
>
> > I have sent the public key to the keyserver as you have suggested.
>
> Great, thanks!  It'll probably take a bit of time for it to propagate.
>
> > Regarding the dependencies you have mentioned, they shouldn't really
> > be such. The tarball includes snapcraft sources for creating artifacts
> > for the snap store (think of this as the analogue of the debian/
> > folder but for another type of repository, https://snapcraft.io/).
> > This tool is then not required for the application to run and to even
> > build.
>
> Yeah, but the build system still calls it.  After your changes at least
> repacking the sources succeeds, but the actual build still fails:
>
>
> Command: dpkg-buildpackage -us -uc -b -rfakeroot
> dpkg-buildpackage: info: source package austin
> dpkg-buildpackage: info: source version 0.6.1~beta-1
> dpkg-buildpackage: info: source distribution unstable
> dpkg-buildpackage: info: source changed by Gabriele N. Tornetta <phoenix1987@gmail.com>
>  dpkg-source --before-build .
> dpkg-buildpackage: info: host architecture amd64
>  fakeroot debian/rules clean
> dh clean
>    dh_auto_clean
>         make -j6 clean
> make[1]: Entering directory '/<<PKGBUILDDIR>>'
> snapcraft clean
> make[1]: snapcraft: Command not found
> make[1]: *** [Makefile:23: clean] Error 127
> make[1]: Leaving directory '/<<PKGBUILDDIR>>'
> dh_auto_clean: make -j6 clean returned exit code 2
>
> > In fact, all that austin requires is a C compiler. Python is
> > only a test dependency as it is invoked during tests. The Makefile
> > included in the tarball is not intended for autogeneration, but just
> > to simplify the testing during development. The package should be
> > built with the standard autoreconf process instead, as I think is
> > currently happening.
>
> Seems like this is the problem:
>
> dpkg-buildpackage: info: source package austin
> dpkg-buildpackage: info: source version 0.6.1~beta-1
> dpkg-buildpackage: info: source distribution unstable
> dpkg-buildpackage: info: source changed by Gabriele N. Tornetta <phoenix1987@gmail.com>
>  dpkg-source --before-build .
>  fakeroot debian/rules clean
> dh clean
>    dh_clean
>  dpkg-source -b .
> dpkg-source: info: using source format '3.0 (quilt)'
> dpkg-source: info: building austin using existing ./austin_0.6.1~beta.orig.tar.gz
> dpkg-source: info: using patch list from debian/patches/series
> dpkg-source: warning: ignoring deletion of file Makefile, use --include-removal to override
> dpkg-source: warning: ignoring deletion of file src/Makefile.in, use --include-removal to override
> dpkg-source: warning: ignoring deletion of file src/austin, use --include-removal to override
> dpkg-source: warning: ignoring deletion of file test/test_process_iter.py, use --include-removal to override
> dpkg-source: info: building austin in austin_0.6.1~beta-1.debian.tar.xz
> dpkg-source: info: building austin in austin_0.6.1~beta-1.dsc
>  dpkg-genbuildinfo --build=source
>  dpkg-genchanges --build=source >../austin_0.6.1~beta-1_source.changes
> dpkg-genchanges: info: including full source code in upload
>  dpkg-source --after-build .
> dpkg-buildpackage: info: full upload (original source is included)
>
> The patch system ignores deletions unless specifically told so (there are
> usually nicer ways to get this effect), which apparently results in some old
> version of the Makefile being used.  Normally, dh_autoreconf would generate
> a proper version, but you specifically disable it in debian/rules.  That's
> bad for other reasons as well -- the package won't see improvements from new
> versions of autotools (bug fixes, support for new architectures, etc).
>
> > Please note that I have some local changes that fix some of the
> > lintian warnings. Could you let me know if I should push them (this is
> > the commit on GH:
> > https://github.com/P403n1x87/austin/commit/8a1b2da53ab7d9fc6d93385677b2f2c75c64a3c9).
>
> Those changes look good -- especially running tests during build is
> something that just can't be recommended too much.  You really don't want to
> release something that compiles but doesn't work.
>
>
> One other comment: python2 is on its way out, with Buster being the last
> release to support it (and even that included lots of grumbling).  It's a
> waste of time to add python2 specific code in a new package, and you'd need
> to remove it shortly anyway.
>
>
> Meow!
> --
> ⢀⣴⠾⠻⢶⣦⠀
> ⣾⠁⢠⠒⠀⣿⡁ Remember, the S in "IoT" stands for Security, while P stands
> ⢿⡄⠘⠷⠚⠋⠀ for Privacy.
> ⠈⠳⣄⠀⠀⠀⠀



-- 
"Egli è scritto in lingua matematica, e i caratteri son triangoli,
cerchi, ed altre figure
geometriche, senza i quali mezzi è impossibile a intenderne umanamente parola;
senza questi è un aggirarsi vanamente per un oscuro laberinto."

-- G. Galilei, Il saggiatore.


Reply to: